JP5115089B2 - Keyword extraction method - Google Patents

Description

  The present invention relates to a technology for accessing a website based on information in an image.

  The mainstream of information (publicity activities) to send information to many people for the purpose of promoting product sales and improving the corporate image (commercial message) broadcasting, advertising in newspapers and magazines. Yes and effective.

  Also, with the spread of the Internet, public relations activities on the website have become important.

  Although TV commercials and newspaper advertisements have the advantage of reaching many people in a timely manner, there is a problem that broadcast time and space are limited and a large amount of information cannot be transmitted.

  On the other hand, although advertisements by websites have the advantage of being able to convey the information desired by users (consumers) in detail, they need to be accessed by users and cannot be accessed by consumers who do not know the existence of the website or information itself (that is, Problem).

  For this reason, when advertising on television and newspapers, there is an attempt to inform the existence of a website and induce people interested in the contents of the advertisement to the website.

  For example, the URL of the website is displayed in the CM to encourage access via the Internet.

  However, since URLs are difficult to remember and there are many input mistakes when accessing them, it has not always been possible to be surely guided to a website.

  Therefore, as shown in FIG. 9, there is also a method of displaying a model similar to a search field of a search engine in which a keyword is input, and similarly inputting a keyword from the browser into the search field to perform a search and prompting to access the website. is there. With this method, keywords that are easy to remember can be used for searching as displayed, and can be easily accessed.

Moreover, as a prior art relevant to this invention, there exists a technique disclosed by the following literature, for example.
JP 2002-290947 A Survey of reactivity to internet-linked TV commercials, Nikkei Business Publications, search date 2007/07/26, http://www.nikkeibp.co.jp/netmarketing/databox/nmdb/061201_crossmedia/

  However, if the search keyword is advertised on TV or newspaper and the Internet is searched, the media that advertises the keyword (TV or newspaper) is different from the media that is searched (Internet). There is a problem that it is necessary to start up a browser and perform a search, and it takes time and effort, so the search motivation is reduced.

Therefore, the present invention extracts a search keyword from an image simulating a search field of a search engine,
Provided is a technique for facilitating web access based on information (search keyword) in an advertisement by searching with the search keyword.

In order to solve the above problems, the present invention employs the following configuration.
That is, the keyword extraction method of the present invention is:
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
Is executed by the computer.

In addition, the search method of the present invention includes:
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
Performing a search process or a search pre-process using the search keyword;
Is executed by the computer.

In addition, the keyword extraction device of the present invention is
An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search region;
Is provided.

Moreover, the search device of the present invention provides
An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search region;
A search processing unit that performs search processing or search pre-processing using the search keyword;
Is provided.

  Further, the present invention may be a program for causing a computer to execute the above method. Furthermore, the present invention may be a program in which this program is recorded on a computer-readable recording medium. The function can be provided by causing the computer to read and execute the program of the recording medium.

  Here, the computer-readable recording medium refers to a recording medium that accumulates information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action and can be read from the computer. . Examples of such a recording medium that can be removed from the computer include a flexible disk, a magneto-optical disk, a CD-ROM, a CD-R / W, a DVD, a DAT, an 8 mm tape, and a memory card.

  Further, there are a hard disk, a ROM (read only memory), and the like as a recording medium fixed to the computer.

  According to the present invention, there is provided a technique for facilitating web access based on information (search keyword) in an advertisement by extracting a search keyword from an image simulating a search field of a search engine and searching with the search keyword. it can.

<Device configuration>
The best mode for carrying out the present invention will be described below with reference to the drawings. The configuration of the following embodiment is an exemplification, and the present invention is not limited to the configuration of the embodiment.
FIG. 1 is a schematic diagram of a search system according to the present embodiment.
The search system 10 of this example is a station that provides information via a network such as a station-side device 101 of a broadcasting station that performs a television broadcast, a user terminal 1 that receives a moving image broadcast from the station-side device 101, and the Internet. The server 2 includes a search server (search engine) 3 that provides a search service for information provided by the web server 2, a ranking server 4, and the like.

The user terminal 1 corresponds to a search device or a search keyword extraction device of the system 10.
The user terminal 1 includes an arithmetic processing unit 12 including a CPU (central processing unit) and a main memory, a storage unit (hard disk) 13 storing data and software for arithmetic processing, an input / output port 14, a communication control unit (CCU). : Communication Control Unit) 15 and the like.

  An input device such as a keyboard, a mouse, a CD-ROM drive, a TV (Television) receiver 16, and an output device such as a display device or a printer are appropriately connected to the input / output port 14.

A TV receiver (tuner) 16 receives radio waves from a broadcasting station via a TV antenna, converts them into electrical signals (image information), and inputs them to the input / output port 14.
The CCU 15 communicates with other computers via a network.

  An operating system (OS) and application software (keyword extraction program, search program) are installed in the storage unit 13.

  The arithmetic processing unit 12 appropriately reads out and executes the OS and application program from the storage unit 13, and performs arithmetic processing on information input from the input / output port 14 and the CCU 15 and information read from the storage unit 13. It also functions as an acquisition unit, analysis unit, extraction unit, playback unit, and instruction reception unit.

  The image acquisition unit acquires image information. For example, image information received by the TV receiving unit 16 is received, or image information stored (recorded) in the storage unit 13 is read and acquired.

  The reproduction unit reproduces the moving image based on the image information acquired by the image acquisition unit. That is, a moving image is displayed on the display unit, and the sound of the moving image is output from the speaker. Further, the playback unit notifies the TV reception unit 16 of a channel to be received or converted in accordance with a user operation or the like for playback.

  The instruction receiving unit receives a search instruction (instruction signal) by a user operation.

  As the analysis unit, a part corresponding to a predetermined pattern imitating a search field is specified as a pseudo search field area in the image information constituting the moving image.

  The extraction unit performs character recognition in the pseudo search area and extracts a search keyword.

  The search processing unit uses the search keyword to perform search processing or search pre-processing. For example, a search request including a search keyword is transmitted to the search server 3 via the CCU 15, and the search result returned from the search server is displayed on the display unit. The search processing unit also has a function of accessing the web server based on search results such as a summary of the displayed content and a hyperlink, and displaying the content. The search processing unit may use a general web browser.

  On the other hand, the search server 3 responds based on means for receiving a search request from the search processing unit (web browser) of the user terminal 1, storage means for storing information on the web server 2, and keywords of the received search request. It is a so-called search engine, which is a general computer having means for searching for information of the web server 2 from storage means, means for transmitting search results to the requesting user terminal 1, and the like.

  The web server 2 is connected to other computers such as the user terminal 1 and the search server 3 via a network such as the Internet. The web server 2 provides (sends) the content (file) specified by the access request (such as URL) from another computer to the requesting computer. The web server 2 has a well-known configuration and will not be described in detail.

  Similarly, the ranking server (keyword providing server) 4 is connected to other computers such as the user terminal 1 and the search server via a network such as the Internet. The ranking server 4 stores ranking information obtained by sorting the keywords used for the search on the search site in the descending order of the number of searches, and provides the keywords (ranking information) in response to requests from other computers. To do. The ranking server 4 may also be used as the search server 3. Further, the keyword used in the CM may be stored in the storage unit by the operator. Since the ranking server 4 has the same configuration as a general web server, detailed description thereof is omitted.

  <retrieval method>

  Next, a search method (including a search keyword extraction method) executed by the user terminal 1 having these configurations based on a search program will be described with reference to FIG.

  As shown in FIG. 2, when the user terminal 1 is instructed to view (reproduce) the TV by the user's operation, the reproduction unit moves the moving image based on the image information received from the storage unit 13 or the TV reception unit. Play.

  At this time, the image acquisition unit of the user terminal 1 specifies (acquires), as a frame to be analyzed (image information), a frame that satisfies a predetermined condition (described later) from among a series of frames constituting the moving image (step 1 and thereafter) Abbreviated as S1).

Next, the analysis unit analyzes the identified frame, and identifies a pseudo search field area corresponding to a predetermined pattern imitating the search field of the search engine (S2).
Further, the extraction unit performs character recognition on the pseudo search field area and extracts keywords (S3).

And a search part starts a web browser, transmits the keyword extracted by the said extraction part to a search server, searches, and displays a search result (S4).
Specific processing of each step will be described using the following specific examples.

<Specific example 1>
FIG. 3 is an explanatory diagram of a method for automatically specifying and searching for a predetermined portion (tail) of a CM from a moving image.
First, the image acquisition unit detects a CM portion other than the main program in the moving image (S11).
This CM detection method is specified as CM when the following conditions are satisfied.
1. When the preceding and succeeding frames are compared, the entire frame is different (the degree of coincidence is less than a predetermined value), that is, when the cut is switched, there is a silent time longer than the predetermined.
2. Since the main part of the program is broadcast in monaural and switched to stereo broadcasting during commercials, the period from when monaural is switched to stereo until it returns to monaural.
3. 3. When the cut is switched in a predetermined time (for example, a multiple of 15 seconds). Predetermined time (for example, 5 minutes before the hour)
5. 5 minutes before and after the program switching time based on the program information acquired by EPG, and the time when the program is equally divided (2 or 4) from the program switching time

  In addition, as a CM detection method, other known methods may be used, or these may be used in combination.

  Next, the image acquisition unit sets a time L that is a reference for the length of the CM. In this example, L = 15 (seconds) is set (S12).

  Note that the timing for notifying the keyword or the like is likely to be the end of the CM, so a predetermined time (L / 2, L / 3, L-5 (seconds) from the beginning of the CM detected based on the above conditions. ) Acquire frames from L to L later. In this example, L-5 to L are set (S13).

  Then, the analysis unit analyzes the frame (image information) acquired by the image acquisition unit, specifies a portion corresponding to a predetermined pattern imitating a search field as a pseudo search field area, and the extraction unit identifies the pseudo search field. Characters are extracted from the area (S14).

  At this time, the analysis unit scans the image in the horizontal direction (main scanning direction) and the vertical direction (sub-scanning direction) in order to specify a region imitating the search field in the image as shown in FIG. Alternatively, a portion that is a straight line continuously longer than a predetermined distance in the vertical direction is extracted. And the area | region where this straight line is a rectangular shape is made into the pseudo search column area | region.

  In particular, in the present embodiment, a predetermined pattern in which a rectangle 62 having a short width (horizontal direction) is adjacent to one rectangle 61 and a character “search” is present in the short rectangle, that is, there is an image corresponding to a search button. If it matches, the portion of the rectangle 61 is specified as the pseudo search field area.

  If the keyword can be extracted at this time, the keyword is notified to the search unit, and if the keyword cannot be extracted, the image acquisition unit is notified that extraction is impossible (S15).

  The image acquisition unit that has received the notification indicating that extraction is not possible determines whether or not the frame targeted for extraction has reached L seconds after the beginning of the CM (S16). The next frame is acquired (S17).

If it is a frame after L seconds in step 16, it is determined whether or not L is less than 60 seconds (S18). If it is not less than 60 seconds, the process ends. In addition (S19), if the maximum value of CM (for example, 60 seconds) is not exceeded, the process returns to step 13 to acquire a frame (S20).

  Note that when acquiring frames in step 13, all frames from L-5 to L seconds may be acquired. However, in the case of MPEG moving images, only I pictures may be acquired. If the configuration is such that only the I picture is acquired, the amount of processing can be reduced.

<Specific example 2>
FIG. 4 is an explanatory diagram of a method for extracting a keyword by specifying a frame in accordance with a user operation, and FIG. 5 is an explanatory diagram of an analysis order of this example.

  First, when a keyword acquisition instruction is input by a user input operation using a keyboard or remote controller connected via the input / output port 14 (S21), the image acquisition unit displays the frame being played back when the input is received. Is acquired (S22).

  Then, the analysis unit analyzes the frame (image information) acquired by the image acquisition unit, specifies a portion corresponding to a predetermined pattern imitating a search field as a pseudo search field area, and the extraction unit identifies the pseudo search field. Characters are extracted from the area (S23).

  If the keyword can be extracted at this time, the keyword is notified to the search unit. If the keyword cannot be extracted, the image acquisition unit is notified that extraction is impossible (S24).

  The image acquisition unit that has received the notification that extraction is not possible determines whether or not the extraction target frame is the previous frame that received the input (S25). It is determined whether the frame is N seconds before the point in time (S26).

  If it is not a frame before N seconds in step 26, the previous frame is acquired (S27), and if it is a frame before N seconds, the next frame at the time of receiving the input is acquired (S28).

  On the other hand, if it is determined in step 25 that the frame has been received, it is determined whether it is a frame after M seconds from the time when the input is received. If not, the image acquisition unit is notified. (S29) The image acquisition unit acquires the next frame (S30). If it is determined in step 29 that the frame is M seconds later, the extraction process ends.

  Thus, when specifying a frame according to the user's input, since the user is likely to perform an input operation after finding a keyword in the moving image, the frame is specified sequentially from the time of receiving the input, If a keyword is not extracted, the analysis processing can be quickly performed by specifying the frame after the time of receiving the input as the analysis target.

<Specific example 3>
FIG. 6 is an explanatory diagram of an example in which keywords are extracted during live broadcasting and the extracted keywords are buffered.

  First, the image acquisition unit determines whether or not the reproduced moving image is a live broadcast (information received by the TV reception unit) (S31), and if it is a live broadcast, acquires the current frame (S32).

The analysis unit specifies a pseudo search field area from the acquired frame (S33). If the pseudo search field area can be specified here, the extraction unit performs character recognition for the pseudo search field area (S34-S35), extracts a keyword, and if the pseudo search field area cannot be specified, step S31. Return to.

  If the keyword can be extracted in step 35, it is determined whether or not the buffer is full (S36-S37). If it is full, the oldest data in the buffer is deleted (S38) and the extracted keyword is added to the buffer (S38). S39).

  For the buffered keyword, for example, when the user gives an instruction, the extraction unit reads the latest keyword from the buffer, notifies the search unit, and performs a search.

  In this example, the keyword extracted in step 35 is stored in the buffer. However, the pseudo search field area specified in step 33 may be stored in the buffer, and steps 34 and 35 may be omitted.

  Further, when frames in live broadcast are sequentially acquired in step 32, all frames constituting a moving image may be acquired. However, in the case of an MPEG moving image, only an I picture may be stored. . Thereby, the storage capacity and the processing amount of analysis can be suppressed.

<Specific Example 4>
FIG. 7 is an explanatory diagram of a method for analyzing a specified frame, extracting a keyword, and searching.
First, when the analysis unit analyzes the frame to be analyzed and identifies the pseudo search field area (S41), the extraction unit performs character recognition of the area (S42).

  When a keyword can be extracted from the pseudo search field (S43), the keyword is compared with a keyword stored in the database (storage unit) of the ranking server 4 to determine whether there is a similar keyword. (S44).

  If there is a similar keyword, the search unit sets the keyword as a search keyword of the web browser, accesses the search site, and performs a search (S45, S47). If there is no similar keyword in step 44, the extracted keyword is set as a search keyword to perform a search (S46, S47).

  When the search site performs a search using the keyword and returns the search result (S48), the search unit of the user terminal 1 displays the highest content of the search result on the display unit (S49).

FIG. 8 is a diagram illustrating a display example of search results.
The address column 51 of the search result window displays the URL of the searched content, and the content column 52 displays the content received from the web server.

  In this example, a search result list 54 and a search keyword 53 are also displayed in a separate frame from this content. If other links are selected from the search result list, contents other than the above can be browsed.

Note that the search result display method is not limited to the above, and only the top-level content or only the search result list may be used.
In addition, without performing a search by keyword, as a pre-search process, the web browser is started, the process is stopped with the extracted keyword inserted in the search field of the search page of the search server 2, and the user's operation is waited. Anyway.

<Others>
The present invention is not limited to the illustrated examples described above, and various modifications can be made without departing from the scope of the present invention.

  For example, in the above-described example, the moving image received by the TV receiving unit or the moving image read from the storage unit is used as the image information. However, the present invention is not limited to this. May be image information obtained by scanning with a scanner.

  Moreover, even if it is the structure added below, the effect similar to the above-mentioned embodiment is acquired. These components can be combined as much as possible.

(Appendix 1)
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
A keyword extraction method executed by a computer.

(Appendix 2)
The keyword extraction method according to appendix 1, wherein frames of a predetermined interval are detected as the image information among a series of frames constituting a moving image.

(Appendix 3)
The keyword extraction method according to supplementary note 1, wherein frames in moving images are compared and a frame satisfying a CM condition is detected as the image information.

(Appendix 4)
The keyword extraction method according to supplementary note 3, wherein a frame within a predetermined period among a plurality of frames that satisfy the CM condition and is a group that is continuous in time series is the analysis target.

(Appendix 5)
The keyword extraction method according to claim 4, wherein the predetermined period is a latter half portion, a portion after a predetermined time from the beginning, or a portion before a predetermined time from the last in the frame group.

(Appendix 6)
Replaying a moving image based on the image information;
Receiving an instruction signal by a user operation; and
6. The keyword extraction method according to any one of appendices 2 to 5, wherein a frame being reproduced when the instruction signal is received or a frame separated from the frame being reproduced for a predetermined time is the target of the analysis.

(Appendix 7)
A frame that is a predetermined time after a frame that is being played back when the instruction signal is received is subject to the analysis, and when a region corresponding to the predetermined pattern does not exist, a frame that is being played back when the instruction signal is received 7. The keyword extraction method according to appendix 6, wherein a frame before a predetermined time is a target of the analysis.

(Appendix 8)
The keyword extraction method according to appendix 6, wherein the image information is stored in a storage unit, and a frame that is a predetermined time before the frame being reproduced is read from the storage unit when the instruction signal is received and is analyzed.

(Appendix 9)
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
Performing a search process or a search pre-process using the search keyword;
A search method that the computer performs.

(Appendix 10)
The search method according to appendix 9, wherein the search preprocessing is a process of starting a browser and setting the search keyword as a search parameter of a search site.

(Appendix 11)
The search method according to appendix 9, wherein the search process is a process of starting a browser and displaying a result of searching the search keyword on a search site on the browser.

(Appendix 12)
An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search region;
Extraction device with.

(Appendix 13)
The keyword extraction device according to supplementary note 12, wherein the image acquisition unit detects frames at a predetermined interval as the image information among a series of frames constituting a moving image.

(Appendix 14)
The keyword extraction device according to attachment 12, wherein the image acquisition unit compares frames in a moving image and detects a frame that satisfies a CM condition as the image information.

(Appendix 15)
The keyword according to supplementary note 14, wherein the analysis unit is configured to analyze a frame within a predetermined period among a plurality of frames that satisfy the CM condition and are a group of time-series continuous groups. Extraction device.

(Appendix 16)
The keyword extracting device according to supplementary note 15, wherein the predetermined period is a latter half portion, a portion after a predetermined time from the beginning, or a portion before a predetermined time from the last in the frame group.

(Appendix 17)
A playback unit for playing back a moving image based on the image information;
An instruction receiving unit that receives an instruction signal by a user operation, and
17. The keyword extraction device according to any one of appendices 13 to 16, wherein a frame being reproduced when receiving the instruction signal or a frame separated from the frame being reproduced for a predetermined time is the target of the analysis.

(Appendix 18)
A frame that is a predetermined time after a frame that is being played back when the instruction signal is received is subject to the analysis, and when a region corresponding to the predetermined pattern does not exist, a frame that is being played back when the instruction signal is received 18. The keyword extracting device according to appendix 17, wherein a frame before a predetermined time is a target of the analysis.

(Appendix 19)
An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search region;
A search processing unit that performs search processing or search pre-processing using the search keyword;
Search device equipped with.

(Appendix 20)
The search device according to appendix 19, wherein the search preprocessing is a process of starting a browser and setting the search keyword as a search parameter of a search site.

(Appendix 21)
The search device according to appendix 19, wherein the search process is a process of starting a browser and displaying a result of searching the search keyword on a search site on the browser.

(Appendix 22)
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
Keyword extraction program that causes a computer to execute.

(Appendix 23)
Item 23. The keyword extraction program according to appendix 22, wherein frames of a predetermined interval are detected as the image information in a series of frames constituting a moving image.

(Appendix 24)
Item 23. The keyword extraction program according to appendix 22, wherein frames in moving images are compared and a frame that satisfies a CM condition is detected as the image information.

(Appendix 25)
25. The keyword extraction program according to appendix 24, wherein a frame within a predetermined period is a target of analysis among a plurality of frames that satisfy the CM condition and are a group of frames that are continuous in time series.

(Appendix 26)
26. The keyword extraction program according to appendix 25, wherein the predetermined period is a latter half portion, a portion after a predetermined time from the beginning, or a portion before a predetermined time from the end of the frame group.

(Appendix 27)
Replaying a moving image based on the image information;
Receiving an instruction signal by a user operation; and
27. The keyword extraction program according to any one of appendices 23 to 26, wherein the frame being reproduced when the instruction signal is received or a frame separated from the frame being reproduced for a predetermined time is the target of the analysis.

(Appendix 28)
A frame that is a predetermined time after a frame that is being played back when the instruction signal is received is subject to the analysis, and when a region corresponding to the predetermined pattern does not exist, a frame that is being played back when the instruction signal is received 28. The keyword extraction program according to appendix 27, wherein a frame before a predetermined time is a target of the analysis.

(Appendix 29)
28. The keyword extraction program according to appendix 27, wherein the image information is stored in a storage unit, and when the instruction signal is received, a frame that is a predetermined time before the frame being reproduced is read from the storage unit and analyzed.

(Appendix 30)
Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search area;
Performing a search process or a search pre-process using the search keyword;
Search program that causes a computer to execute.

(Appendix 31)
The search program according to supplementary note 30, wherein the search preprocessing is a process of starting a browser and setting the search keyword as a search parameter of a search site.

(Appendix 32)
The search program according to appendix 30, wherein the search process is a process of starting a browser and displaying a result of searching the search keyword on a search site on the browser.

Schematic diagram of the search system Explanatory diagram of search method (including search keyword extraction method) Explanatory drawing of the method of automatically specifying and searching for a predetermined part (tail) of CM Explanatory drawing of the method of identifying a frame and extracting a keyword according to a user's operation Illustration of analysis order Illustration of buffering keywords extracted during live broadcast Explanatory diagram of how to analyze the frame after identification, extract keywords, and search Figure showing a display example of search results Explanatory drawing of an image having an area simulating a search field

Explanation of symbols

10 Search System 1 User Terminal 2 Web Server 3 Search Server (Search Engine)
4 Ranking server 101 Station side device

Claims (10)

Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search field area;
A keyword extraction method executed by a computer. The keyword extracting method according to claim 1, wherein in the step of acquiring the image information, frames at a predetermined interval are detected as the image information among a series of frames constituting a moving image. The keyword extraction method according to claim 1, wherein in the step of acquiring the image information, frames in moving images are compared, and a frame that satisfies a CM condition is detected as the image information. 4. The analysis of the image information, wherein a frame within a predetermined period is selected as a target of analysis among a plurality of frames that satisfy the CM condition and are a group of time-series continuous groups. Keyword extraction method. 5. The keyword extraction method according to claim 4 , wherein in the analysis of the image information, the predetermined period is a second half portion, a portion after a predetermined time from the beginning, or a portion before a predetermined time from the last in the frame group. Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search field area;
Performing a search process or a search pre-process using the search keyword;
A search method that the computer performs. An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search field area;
A keyword extraction device comprising: An image acquisition unit for acquiring image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
An extraction unit for extracting a search keyword from the pseudo search field area;
A search processing unit that performs search processing or search pre-processing using the search keyword;
Search device equipped with. Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search field area;
Keyword extraction program that causes a computer to execute. Obtaining image information;
Analyzing the image information and identifying a pseudo search field area corresponding to a predetermined pattern imitating a search field;
Extracting a search keyword from the pseudo search field area;
Performing a search process or a search pre-process using the search keyword;
Search program that causes a computer to execute.

Images